A Predictive Model for Emotion Recognition Based on Individual Characteristics and Autonomic Changes

Abstract

Introduction: The importance of individual differences in the problem of emotion recognition has been repeatedly stated in the studies. The major concentration of this study was the prediction of heart rate variability (HRV) changes due to affective stimuli from the subject characteristics. These features were age (A), gender (G), linguality (L), and sleep (S) information. In addition, the most potent combination of individual variables (like gender and age (GA) or age, linguality, and sleep (ALS)) in the estimation of emotional HRV was explored. Methods: To this end, HRV indices of 47 college students exposed to images with four emotional categories, including happy, sad, afraid, and relaxed were analyzed. Then, a novel predictive model was introduced based on the regression equation. Results: The results showed distinctive emotional situations provoke the importance of different individual variable combinations. Generally, the most satisfactory variable arrangement in the prediction of HRV changes due to affective provocations was LS, GL, GA, ALS, and GALS. However, considering each subject separately, these combinations were changed. Conclusion: In conclusion, the suggested simple model is effective in offering new insight into the emotion studies regarding subject characteristics and autonomic parameters.